电力需求侧管理2026，Vol.28Issue(2)：57-63,7.DOI:10.3969/j.issn.1009-1831.2026.02.009
基于CNN-LSTM-CMA-GRU的多尺度中期负荷预测方法
Multi-scale medium-term load forecasting method based on CNN-LSTM-CMA-GRU
摘要
Abstract
Accurate mid-term power load forecasting is crucial for power dispatch and resource optimization.Addressing the practical need for daily peak/valley load management in power scheduling,medium-term forecasting is studied with daily maximum/minimum load as the prediction granularity.To overcome the error accumulation caused by the decay of coupling relationships between historical loads and multi-dimensional external variables in traditional methods,a deep neural network time-series forecasting approach incorporating a cross multi-head attention(CMA)mechanism is proposed.The model features three innovative designs:first,a dual-branch convolutional neural net-work(CNN)and long short-term memory(LSTM)network are used to extract the local pattern features of the load sequence and the global temporal correlation of auxiliary variables;second,a cross-multi-head attention layer is designed to establish a dynamic weight mapping be-tween historical load and external variables in future periods;finally,a gated recurrent unit(GRU)achieves adaptive fusion of multi-scale features.Experimental results demonstrate that the model achieves high accuracy and ro-bustness in power load forecasting tasks.关键词
中期负荷预测/交叉多头注意力/多时间尺度/CNN-LSTM/深度神经网络Key words
medium-term load forecasting/cross-multi-head attention/multi-time scale/CNN-LSTM/deep neural network分类
信息技术与安全科学引用本文复制引用
曹雯,范冰,徐铭铭,景力涛,李德军,汤文俊..基于CNN-LSTM-CMA-GRU的多尺度中期负荷预测方法[J].电力需求侧管理,2026,28(2):57-63,7.基金项目
国家电网有限公司总部科技项目(SGHADK00PJJS2200050) （SGHADK00PJJS2200050）
